Doing research on ‘management fashions’: methodological challenges and opportunities
Published: Dec. 1, 2013
Management fashion theory is a growing research area in management studies. The focus of this management fashion literature is to understand why some management concepts spread quickly and widely, while others do not. However, doing research on fashionable management concepts is a difficult task, and many commentators have pointed out the limitations of the research methods used in extant research. A consequence of these difficulties is that the theory has many understudied areas and ‘blind spots’. This paper aims at providing a review of the research methods typically used in management fashion research, and assessing the strengths and weaknesses of various approaches. Based on this review, the paper suggests research strategies that can be used to illuminate the blind spots of the management fashion theory. The paper distinguishes between research strategies at four analytical levels: the managerial level, the intra-organizational level, the field-level and the cross-national level
Management concepts, research opportunities, research methods, research challenges, management fashions
The importance of research on ‘management fashions’. ‘Management fashion’ emerged as an important research topic in the early 1990s as management scholars sought to explain the continuous launching of new concepts, techniques and buzzwords in the management community (e.g. Abrahamson, 1991, 1996; Gill & Whittle, 1993; Nohria & Eccles, 1992). These researchers focused on how the popularity of management concepts is driven by fashion-setters who influence what is considered ‘fashionable’ in the market place for management knowledge and ideas. Over the course of the last two decades, management fashion theory has grown as a research area and has branched out in different directions (e.g. Abrahamson, 1996; Abrahamson & Eisenman, 2008; Benders & Van Veen, 2001; Carson, Lanier, Carson & Guidry, 2000; Clark, 2004a; Collins, 2012; David & Strang, 2006; Jackson, 2001; Jung & Lieser, 2012; Lieser, 1997; Klincewicz, 2006; Newell, Robertson & Swan, 2001; Scarbrough & Swan, 2001; Spell, 2001; Staw & Epstein, 2000; Swan, 2004; Ten Bos & Heusinkveld, 2008). However, a red thread running through most of this research is the dissemination and diffusion of management concepts, i.e. the processes in which concepts become popular and ‘fashionable’.
Management concepts and management fashions. Before proceeding, it is necessary to clarify the difference between management concepts and management fashions. Management concepts can be seen as “prescriptive, more or less coherent views on management” (Benders & Verlaar, 2003, p. 758). Fashionable management concepts, or ‘management fashions’, are those “management concepts that relatively speedily gain large shares in the public management discourse” (Jung & Lieser, 2012, p. 329). Thus, fashionable management concepts constitute a sub-set of the total supply of management concepts as not all management concepts succeed in becoming popular (Braarn, Heusinkveld, Benders & Aubel, 2002, p. 4). In some cases there is never a wave of interest, and the new concepts instead die off shortly after their introduction (Benders, van den Berg & van Bijsterveld, 1998). It follows that a central research question for management fashion researchers is why some management concepts spread quickly and widely and ‘stick around’, while others do not.
Indeed, researchers have studied the diffusion and popularization of a wide range of management concepts such as Activity-Based Costing (Mahni, 1999), Lean (Benders & Van Bijsterveld, 2000), Business Process Reengineering (Benders et al., 1998; Heusinkveld & Benders, 2001), Talent Management (Iles, Preece, & Chuai, 2010; Preece, Iles, & Chuai, 2011), Quality Management (David & Strang, 2006; Thawesaengskulthai & Tannock, 2008), Knowledge Management (Grant, 2011; Klincewicz, 2006; Scarbrough & Swan, 2001), and Balanced Scorecards (Ax & Bjomenak, 2005; Braarn, Benders, & Heusinkveld, 2007; Madsen & Slätten, 2013). These studies have shown the importance of fashion-setting actors on the supply-side (in particular, consultants and ‘management gurus’) in driving the diffusion and institutionalization of fashions. In addition, these studies have shown that the content of fashions varies across time and space as actors interpret and enact fashionable concepts in various ways (cf. Benders & Van Veen, 2001; Giroux, 2006). These findings have several implications for empirical research on management fashions, as studying the supply-side (or discourse surrounding a concept) only tells part of the story. In order to obtain a ‘balanced’ overview of a fashion’s impact, it is also necessary to study how the concept is used on the demand-side (Benders & Van Veen, 2001; Clark, 2004a; Madsen, 2011).
The evolution of the management fashion research literature. As pointed out earlier, the management fashion research literature has branched out in different directions. Viewed as a whole, the focus of management fashion research has shifted somewhat over time, from a focus on the macro-level dynamics of fashion markets to the micro-level interpretations and applications of fashions by users or ‘consumers’. As pointed out by Perkmann and Spicer (2008) the initial focus of management fashion research was on understanding the workings of the management fashion market, such as its structure and dynamics. The seminal papers in the management fashion tradition utilized print-media indicators (PMIs) from large article databases to trace the lifecycle and evolution of fashions over time (Abrahamson, 1996; Abrahamson & Fairchild, 1999; Carson et al., 2000). Much of the empirical management fashion research that followed in the footsteps of these early papers also utilized primarily quantitative macro-level data. Modell (2009, p. 60) points out that a “majority of empirical research following the fad and fashion perspectives has focused on broad diffusion and adoption patterns and relies heavily on survey data or secondary sources. ” Although such quantitative research has been fruitful for understanding the macro-aspects of the life-cycle and evolution of fashions, commentators have noted that other research methods (e.g. interviews and observations) are needed to be able to say something about how fashions are interpreted (Benders & Van Bijsterveld, 2000; Benders & Van Veen, 2001) and ‘consumed’ (Heusinkveld, Sturdy & Werr, 2011; Wilhelm & Bort, 2012), particularly at the intra- organizational (Heusinkveld & Benders, 2012) and managerial levels (van Rossem & van Veen, 2011).
Purpose and contribution. Against this background, the purpose of this paper is to discuss methodological challenges and opportunities in management fashion research. The paper contributes to the existing management fashion literature in at least two ways. First, the paper provides a comprehensive review of the research methods employed in the extant management fashion literature. Second, the paper outlines specific research strategies which can be employed in future studies of fashionable management concepts. These strategies address some of the research gaps and blind spots in the management fashion research tradition.
The remainder of the paper is structured as follows. Section 1 presents a review of the research methods used in the existing research literature, and a discussion of the strengths and weaknesses of the respective methods. Section 2 of the paper identifies key methodological challenges typically encountered by researchers studying management fashions. In section 3 we propose some research strategies that can be used in future management fashion studies. Here, we distinguish between the strategies that can be used at each analytical level, from the international level to the level of the individual manager. The paper ends with a conclusion.
In the previous section we made a broad distinction between micro- and macro-oriented research on management fashions. In this section we discuss more in detail the various research methods that are typically used in micro- and macro-oriented management fashion research. We focus on the four research methods most commonly used in management fashion research, starting with PMIs and surveys, two methods that are commonly used to study the macro-level impact of fashions. This is followed by a discussion of the use of interviews and observations, two qualitative research methods which can shed light on the micro-level impact of fashions. Table 1 summarizes the strengths and weaknesses of the four research methods.
1.1. Print-media indicators. Using print-media indicators (PMI) involves tracing the popularity of management concepts by using archival data from electronic article databases such as ABI/Inform or ProQuest (Benders, Nijholt & Heusinkveld, 2007). PMI research is a well-established method in management fashion research, having been extensively used in the seminal studies in the management fashion tradition (e.g. Abrahamson & Fairchild, 1999; Carson et al., 2000). PMI research has been fruitful for our understanding of the macro-level structure and dynamics of the management fashion market (Perkmann & Spicer, 2008). For example, PMI research has highlighted the role of the business media in the dissemination of management fashions (e.g. Alvarez, Mazza & Pedersen, 2005; Mazza & Alvarez, 2000; Ruling, 2005). Such research has also helped our understanding of the various types of actors (e.g. consultants, gurus and academics) that are contributing to the public discourse about concepts in the business media, and shaping the evolution of a fashion in a particular community (Braarn et al., 2007; Heusinkveld, 2004). Thus, PMIs can be useful for establishing broader macro-patterns in the impact and evolution of a concept in different countries, or in different professional or social communities within one country. In particular, PMIs provide valuable data about the role and activities of the supply-side of fashions.
Besides, using PMIs has many advantages of a more practical nature (Benders et al., 2007), which is important since studying fashionable management concepts is time- and resource-consuming (Braarn et al., 2007). For example, it is relatively easy for researchers to get large amounts of PMI data from international databases. Most industrialized countries have well-developed databases with full-text archives of national publications such as newspapers, professional magazines, and academic journals. Thus, it is a cost-effective way of gathering large amounts of data about the public discourse surrounding a concept.
The PMI data can also be used in different ways. For example, the data can be used to analyze the evolution of a concept at the international level, e.g. comparisons between different regions or groups of countries. In addition to quantitative analyses (e.g. article counts over time), the data can also be analyzed qualitatively using techniques such as content analysis (e.g. Krippendorff, 2012) or discourse analysis (e.g. Gee, 2010). For example, researchers can analyze the rhetoric and language used in articles, and whether the authors have a positive, neutral or negative stance towards the concept. It is also possible to track the backgrounds of the authors, and to analyze the types of actors around providing the discourse, e.g. consultants, academics or journalists (Heusinkveld, 2004, p. 31).
However, commentators have pointed out that PMI- research has several limitations (Benders et al., 2007; Benders & Van Veen, 2001; Clark, 2004a; Nijholt & Benders, 2007; Örtenblad, 2010; Perkmann & Spicer, 2008; Rovik, 2007,2011), and that such data are not a good proxy for a concept’s actual impact in organizational practice. Benders and Van Veen (2001) point out that the discourse surrounding a fashionable concept and the actual organizational changes associated with this discourse are often loosely coupled. The fact that a concept is subject to widespread attention in the media does not necessarily mean that the concept is widely used in practice (Benders & Van Veen, 2001; Clark, 2004a), and some fashions might still be used even if they are not much talked about anymore (Perkmann & Spicer, 2008). For example, Nijholt and Benders (2007) note that much of the research using PMIs assumes that the media discourse co-evolves with actual usage in practice, and that this assumption does not always hold true. Instead, management researchers increasingly are taking the position that fashions may ‘stick around’ as more persistent and permanent (i.e. institu-tionalized) practices even long after the media interest and initial ‘hype’ in the business discourse has waned (Heusinkveld & Benders, 2012; Perkmann & Spicer, 2008; Rovik, 2011).
Thus, it can be argued that the PMI method offers only a partial and ‘one-sided’ picture of the impact of a fashionable concept. While it provides valuable insight into the structure and dynamics of the supply-side of a particular concept, including the types of actors supplying this discourse, PMIs offers little insight into a concept’s impact on the demandside of the market (Benders & Van Veen, 2001; Clark, 2004a). Hence, PMIs are not appropriate for addressing research questions such as how management fashions are ‘consumed’ (Heusinkveld et al., 2011), and how they are implemented and evolve as practices in organizations (Heusinkveld & Benders, 2012; Heusinkveld, Benders & Hillebrand, 2013; Rovik, 2011).
1.2. Surveys have also been used to some extent in previous management fashion research, particularly to map the diffusion of concepts in a population of demand-side organizations. Examples of surveys include Malmi’s (1999) study of the diffusion of the ABC concept, and Bain & Company’s biannual longitudinal study of managers’ use of management tools (e.g. Rigby & Bilodeau, 2007; Rigby & Bilodeau, 2009; Rigby & Bilodeau, 2011; Rigby & Bilodeau, 2013).
Management fashion researchers have pointed out that surveys have obvious advantages when it comes to getting a broader overview of a concept’s impact in a larger population of organizations (Benders & Van Bijsterveld, 2000). Since organizations are not required to report their use of management concepts, it is necessary to ask them directly, and a survey is the only feasible way of obtaining such data from a large population of organizations. Like PMIs, surveys also have certain practical advantages. Since the arrival of electronic survey programs, the financial cost of sending out surveys is virtually zero (Jansen, Corley & Jansen, 2007; Sue & Ritter, 2011). Consequently, surveys are a cost-effective way of gathering data about management fashions.
The flip side of this is that organizations and managers are bombarded with electronic surveys from researchers, students, and other organizations (e.g. consulting firms and professional organizations). This makes getting a high response rate more difficult than ever. The perhaps biggest drawback lies in the limitations of the survey method in obtaining valid data about the actual usage of management fashions. By asking survey respondents about the adoption of a management concept, you may obtain information about whether the management concept is adopted, but not how or in what way it is adopted (Benders & Van Bijsterveld, 2000). Organizations may implement a concept in various ways which makes survey research highly problematic since it usually has to treat adoption as a dichotomous, i.e. either-or, variable (Rovik, 2011). The aforementioned global longitudinal survey conducted by Bain & Company, while providing a very useful indication of macrotrends in the popularity of management concept, suffers from these problems, as we know little about what is adopted under the different labels.
Despite these drawbacks, surveys have an important role in certain types of management fashion research since they are necessary in order to obtain a broad overview of a concept’s impact in a population of organizations (Benders & Van Bijsterveld, 2000; Braarn et al., 2007).
1.3. Qualitative interviews. Qualitative interviews have been used in previous management fashion studies to explore how fashions have been interpreted and implemented by actors such as consultants and users of the concepts (Benders & Van Bijsterveld, 2000; Benders et al., 1998; Braarn et al., 2007; Heusinkveld, 2004). In comparison with surveys, one of the advantages of qualitative interviews is that they can provide more in-depth insight into how organizations have interpreted and implemented management concepts. On the other hand, qualitative interviews are less useful for assessing the impact of a concept in a large population of organizations.
When using the interview method, the researcher has to decide whether to focus on interviewing several informants in one organization (i.e. depth) or interviewing informants in several organizations (i.e. breadth). If interviewing one informant from many different organizations, die research assumes the form of a “qualitative survey” (Jansen, 2010). The goal of a qualitative survey is still not generalization in a statistical sense of the word, but to explore the extent to which the interpretations and experiences are shared across a larger number of organizations. For example, qualitative surveys of multiple organizations could be very useful to supplement and interpret the findings from quantitative survey material.
1.4. Observations. Observational methods have not been much used in management fashion research. As discussed, researchers have traditionally relied heavily on PMI data as a proxy for usage of fashions, or used surveys (or in some cases interviews) as these are often less time- and resource-intensive than direct observation.
This is troubling since many studies have documented a gap between rhetoric and action in organizations (e.g. Meyer & Rowan, 1977; Zbaracki, 1998). This gap might be understated if researchers rely solely on research methods such as surveys or interviews, where the informant may misrepresent the extent to which the concept is used in the organization. By using observational methods, the researchers would be able to get an understanding of how a concept is used and how it actually impacts work practices.
On the other hand, observational methods have certain disadvantages. The biggest drawback is that fieldwork is very time- and resource-intensive. This makes it difficult, or maybe even impossible, to study the impact of a concept across several organizations. In addition, the researcher has to deal with typical problems of getting access. Many organizations are reluctant when it comes to letting researchers spend a significant amount of time in their organization studying the use of management concepts which often are related to sensitive issues such as competitive strategies.
As a whole, however, observational methods have a key function as a way to get a better understanding of the impact of fashions at the micro-level, both in terms of how they impact work practices in the organization, but also how they affect the behavior of groups and individuals.
From the discussion in section 1, it becomes clear that doing empirical research on management fashions poses a number of challenges for researchers. In this section we discuss some key challenges that researchers tend to face in empirical studies of management fashions. Most of these challenges are related to the fact that fashionable management concepts are ‘ideational innovations’ (Benders & Van Veen, 2001), which can be interpreted and implemented in various ways. This makes fashionable management concepts difficult to study using traditional desk research methods (e.g. surveys). Instead, researchers often have to venture out ‘into the wild’ to interview and observe the usage of these concepts in practice, which may be methodologically difficult and costly (Benders et al., 2007). A third issue concerns identifying organizations, informants and obtaining access. This is generally a bigger hurdle when utilizing time- and resource-intensive methods such as interviews and observations. These issues are elaborated below.
2.1. The ideational nature of management concepts. Several researchers have noted that management concepts as administrative innovations are ideational and lacking a material component (Benders & Van Veen, 2001; Fincham & Roslender, 2004). Due to their ‘interpretative space’, management concepts are open to numerous interpretations when implemented in practice (Benders & Van Veen, 2001; Clark, 2004a; Giroux, 2006). Studying the prevalence of material innovations is straightforward since the researcher can just count the number of users (Benders & Van Bijsterveld, 2000). This is not possible in the case of mana-gement concepts as they can be interpreted and used in a variety of ways.
However, as Rovik (2011) points out, much of the research on management fashions still tends to treat adoption as a dichotomous variable (i.e. adoption or not). This is problematic, and instead researchers should recognize that fashionable concepts can be adopted, implemented and handled in various ways. However, this opens up a range of issues that researchers have to tackle as they attempt to study these concepts in organizational practice.
2.3. Time- and resource-intensive. As mentioned, research on the implementation of management fashions in organizational practice is rare (Clark, 2004a; Rovik, 2011). This might be due to the fact that such data are difficult and costly to obtain (Benders et al., 2007). These challenges might explain "... the notable lack of studies providing data on the prevalence of fashionable organization concepts... ” (Nijholt & Benders, 2007, p. 637). As discussed in section 1, from a practical stand-point it is arguably easier to administer and carry out database searches and survey research than interviews and observations, as the latter are usually very time- and resource-intensive. In addition, qualitative in-depth investigations also run into other types of challenges which are elaborated below.
2.4. Identifying organizations, informants and obtaining access. Identifying informants is a key challenge in research on fashionable management concepts. It is vital to be able to identify the person(s) in the organizations that is (are) best able to provide insight into how the concept is used in the organization. In some cases, it might be necessary to interview several informants from an organization. However, being able to identify these informants is of little use if one does not succeed in getting access to them. In larger interview studies or multi-case studies the researcher need to deal with not only gatekeepers in one organization, but numerous gatekeepers. A related issue is the fact that such individuals tend to be high-status individuals (or ‘elites’) in the organization, e.g. CEOs, CFOs or project managers. These individuals tend to be busy and have little time to answer surveys or participate in lengthy interviews (cf. Odendahl & Shaw, 2002).
In this section we identify important research questions at each of the four analytical levels, and suggest research strategies that can be used to answer these questions.
3.1. Managerial level. The managerial level is arguably the most overlooked level of analysis in previous research on management fashions. Most of the research has focused on the diffusion of fashions at the national or inter-organizational level, and little work has looked at the role of managers in the adoption and implementation of these fashionable concepts. Several researchers have noted that we know little about how managers encounter, become aware of, and develop interest in new management concepts (Alvarez et al., 2005; Heusinkveld et al., 2011; Powell, Gamrnal & Simard, 2005; van Rossem & van Veen, 2011; van Veen, Bezemer & Karsten, 2011). As pointed out by van Veen et al. (2011, p. 161hardly any theoretical and empirical -work has been done on how managers get to know these concepts, and for what reasons they develop more or less interest in a specific concept (and not others). ” Hence, since there is little previous research to build on and few guidelines in the literature as to how this should be studied, the researcher needs to develop new research strategies. There are, however, based on insights from related fields, many issues that management fashion researchers could explore.
For instance, at the managerial level, researchers of management fashions could explore how individual managers perceive and evaluate fashions, e.g. the process leading up to the adoption and implementation of concepts (van Rossem & van Veen, 2011). For such research, different methods might be used, e.g. in-depth interviews with managers about their thoughts and decision-making process. Another possibility would be to observe internal meetings where the adoption and implementation of concepts are discussed, and meetings with actors from the management knowledge market such as consultants and software vendors.
In line with the so-called ‘practice-turn’ in strategy and accounting research (Vaara & Whittington, 2012; Whittington, 2011), it would also be interesting to explore how fashionable management concepts impact work practices of managers, and are used in day-to- day strategy, accounting, or general managerial work. For this purpose, an in-depth micro-level investigation would be necessary, particularly direct observation of the daily activities of managers, e.g. meetings and conference calls.
3.2. Intra-organizational level. At the intra- organizational level, the focus turns to how fashionable concepts impact practices in organizations, e.g. at the group-level. Important research questions at this level are how the adoption and implementation of concepts affect practices in organizations, and how concepts evolve as practices within organizations (Heusinkveld & Benders, 2012). What types of actors are involved in the institutionalization process where concepts are entrenched (e.g. consultants), and what types of ‘institutional work’ do these actors carry out to ensure that the fashionable concept gains a degree of permanence in the organization? (Perkmann & Spicer, 2008)
At the intra-organizational level, suitable research methods include qualitative methods such as interviews and observations. For instance, interviews with managers and employees at different levels of the organization could provide insights into how a concept has impacted the work practices in different parts of the organization. Observational methods would also be useful at the intra-organizational level. Such research could provide insight into how concepts actually are used inside the organization, e.g. in meetings, in formal and informal communication between different groups and parties, and in decision-making processes.
3.3. Field-level. At the field-level, management fashion researchers seek to understand the impact of fashions at the national or sector level. Here researchers seek to ‘paint a picture’ of the impact of the concept in a population of organizations (Nijholt & Benders, 2007, p. 649) or a ‘reception pattem’ (Braarn et al., 2002). It is often necessary to pattemize and look for general trends in a population of organizations, as it can be difficult, if not impossible, to get a ‘complete’ overview of the impact of the concept at the fieldlevel. For this purpose, more macro-oriented research methods such as surveys and PMIs, preferably in combination, would be appropriate choices of research methods.
Alternatively, a series of interviews with important actors in the field, so-called field experts (Braarn et al., 2002) can be valuable to provide deeper insights into how the concept is typically used in the population. Actors such as consultants and academics involved in the local market may have a longitudinal overview of how the concept has emerged and evolved in the local context. Such interviews can also give more insight into the role of field-level actors involved in the local ‘management fashion arena’ (Clark, 2004b; Jung & Kieser, 2012; Klincewicz, 2006), such as consulting firms, professional organizations, software firms and business school professors. These actors’ activities can be traced and analyzed, and may give important insights into how a concept is interpreted and used in a population of organizations.
3.4. International level. At the international level, one important question is related to the cross-national impact of concepts. Does the impact of concepts vary across different national settings? Could ‘national reception patterns’ (Benders & Van Bijsterveld, 2000) be identified? Abrahamson (1996) pointed out that there may be international differences in the popularity of fashionable concepts, but such questions have rarely been addressed empirically (see e.g. Newell et al., 2001). Other important questions at the international level are the role of institutional (Scott, 2001) and cultural (Hofstede, 1983) differences in the diffusion of management fashions. Well-designed comparative studies could possibly distill the influence of institutional and cultural factors.
The lack of cross-national studies of fashions could be explained by the fact that cross-national studies are hard to administer. There are many challenges with respect to data collection in a cross-national study. In international research projects the data are often gathered by different researchers, for different purposes, and at different points in time, which makes cross-national comparisons difficult.
Hence, an international research project on the impact management fashions should, ideally, combine several different research methods. First of all, the study should utilize macro-oriented research methods such as PMIs and surveys to obtain an overview of the local evolution of the concept. In addition, the study should utilize qualitative methods such as interviews with key country-specific actors to get more in-depth information about the local impact of the concept. Lastly, it would be important to utilize a limited number of case studies in each country. For example, one could select a few cases in each country, and let the cases vary across different dimensions. These case studies would be useful for balancing the macro-oriented data, and distilling country-specific patterns in the interpretations and use of concepts.
To date, management fashion theorists have focused mostly on the macro-level impact of management concepts (Heusinkveld & Benders, 2012; Perkmann & Spicer, 2008). This type of research has mostly been carried out using methods such as print-media citation analyses, or quantitative surveys of the usage of a concept in a population of organizations. Considerably less research has been carried out at the micro-level. For example, few studies have looked at intra-organizational implementation of fashionable management concepts, and how fashions evolve as organizational practices (Clark, 2004a; Heusinkveld & Benders, 2012; Rovik, 2011). Similarly, little research has been carried out at the level of individual managers, e.g. studying how managers encounter, learn about and develop interest in concepts, and their roles in the fashion-setting process (van Rossem & van Veen, 2011; van Veen et al., 2011).
This review article suggests several research questions that should be explored by management fashion researchers. Moreover, we have outlined possible research strategies that could be used to answer these questions. However, we have also noted how the study of management fashions presents researchers with challenges of both theoretical and practical nature. Generally, it is relatively easy to design these studies in theory, but carrying them out in practice is very difficult due to time and resource constraints. This may also explain why there are so many largely unanswered calls for research on how management fashion is implemented in organizations (e.g. Benders & Van Bijsterveld, 2000; Benders & Van Veen, 2001; Clark, 2004a). Still to this day there are hardly any researchers that have stepped up to the plate and actually gathered such data.